Time to change PGN as protocol interchange

[lukasmonk]

I think it's time to change the way in which the chess games are broadcast.

Time to change the protocol, the old standard PGN (valid visually), to other more accurate and more flexible for exchange between programs. A public standard way to interchange chess games.

One idea might be to use JSON or similar(I don´t know much about it), with some fixed fields set as standard, and the ability to add other non-standard fields, but whose information is not lost when using different GUIs.

Some idea about this ?

Cheers
Lucas


Perhaps this is an old topic, but I have not seen posts relating this.

[ZirconiumX]

I think it's time to change the way in which the chess games are broadcast.

Time to change the protocol, the old standard PGN (valid visually), to other more accurate and more flexible for exchange between programs. A public standard way to interchange chess games.

One idea might be to use JSON or similar(I don´t know much about it), with some fixed fields set as standard, and the ability to add other non-standard fields, but whose information is not lost when using different GUIs.

Some idea about this ?

Cheers
Lucas


Perhaps this is an old topic, but I have not seen posts relating this.

What exactly is the problem with PGN?

PGN is flexible (you can add fields all you like), and accurate (it records the game information)

Matthew:out

[hgm]

PGN is awful, because it uses SAN for the moves. To interpret SAN requires knowledge of the game rules. For broadcasting games in WinBoard I already switched to a different format, based on long-algebraic notation.

[Edmund]

PGNs are built for human readers and not for computer processing. In my opinion the most annoying aspects are:

• - SAN are expensive to parse
  - comments not well defined (e.g. for engine output)

[michiguel]

PGNs are built for human readers and not for computer processing. In my opinion the most annoying aspects are:

• - SAN are expensive to parse
  - comments not well defined (e.g. for engine output)

It is for both, humans and computers, and their interaction. It is as good as it gets for what it was designed for, a common and universal language. Of course, you have to make concessions to support carbon and silicon brains. It is not supposed to be the fastest, it is supposed to be understandable and easily used with text tools.

Miguel

[lukasmonk]

From a GUI point of view, (english) PGN is not enough, if you want to add an image to a move, or a link,or a video,or a doc..., the kind of information that is common nowadays.

ChessBase has its own solution: cbv files, but this is privative, IMHO a standard public protocol other than PGN is necessary, to do more easy to transmit all this information between GUIs.

[syzygy]

PGN is awful, because it uses SAN for the moves. To interpret SAN requires knowledge of the game rules. For broadcasting games in WinBoard I already switched to a different format, based on long-algebraic notation.

The rules of chess are known to all.

For other games other formats might be more suitable.

[jdart]

Computers only get faster. Processing a million games typically takes a few minutes, for me. So the processing overhead is not much, IMO.

It is true there are some gray areas not really standardized. Chesbase uses [%eval] and [%emt] in comments for eval and search time (and is annoying about re-writing comments with numeric values into this format). Winboard writes comments like {-0.22/28 1.06}. So yeah, that is a legitimate complaint but it could be solved if tool authors would agree on something.

--Jon

[lukasmonk]

I think it is necessary to distinguish between two aspects of the subject:
1.- the way that games are showed to humans, where I don´t know an alternative to PGN.
2.- and the way the information is saved in the machine, where each program uses its own solutions, where [%eval] in Chessbase is only a patch, as this information is something more/different than a simple comment.I think it is common sense that sooner or later will be improved in this respect.

[hgm]

PGN is awful, because it uses SAN for the moves. To interpret SAN requires knowledge of the game rules. For broadcasting games in WinBoard I already switched to a different format, based on long-algebraic notation.

The rules of chess are known to all.

Not to i386.
Not to C++.
Not to JavaScript.